Inspiration
Stock trading track
What it does
Provides highly customizable models which perform differently depending on time period and features chosen
How we built it
We used python to extract data from the yahoo finance library, then trained multiple different types of machine learning models on it.
Challenges we ran into
It was a little difficult to store some of the models, so we decided to use MongoDB Atlas
Accomplishments that we're proud of
I think the original graph is pretty incredibly. Obviously it's just backtesting but our val_loss was around 0.5, which means that the model we used there was not overfitted
What we learned
Quantitative finance is pretty fun, but difficult to implement over a weekend to show proof
What's next for Machine Learning Model to Teach You How to Trade Stocks
I think we might, after further optimization, end up throwing a little cash in our model just to see if it works or not.
Log in or sign up for Devpost to join the conversation.